Dangers of Using Statistics Wrongly in Scientific Research

Click For Summary

Discussion Overview

The discussion revolves around the misuse of statistics in scientific research, particularly focusing on concepts like p-hacking and the importance of hypothesis-driven analysis. Participants explore the implications of data interpretation, the dangers of drawing conclusions from correlations, and the balance between exploratory data analysis and rigorous statistical methods.

Discussion Character

  • Debate/contested
  • Technical explanation
  • Conceptual clarification

Main Points Raised

  • Some participants emphasize the need for a hypothesis before analyzing data, arguing that data should not be mined for interesting findings without a guiding hypothesis.
  • Others reference articles and blog posts discussing p-hacking and its implications, highlighting examples of misleading correlations, such as the stork population and human population correlation.
  • A participant questions the apparent contradiction between the advice to avoid p-hacking and the recommendation to graph data for visual insights, suggesting that this may involve subjective interpretations of patterns.
  • Some participants propose that analyzing data through multiple methods can provide more robust conclusions, contrasting this with p-hacking practices.
  • There are mentions of data mining practices that resemble p-hacking, where researchers must justify their findings, and examples of successful data analysis that led to the discovery of equations without a clear theoretical basis.

Areas of Agreement / Disagreement

Participants express a range of views on the appropriate methods for data analysis, with some advocating for hypothesis-driven approaches while others highlight the value of exploratory analysis. The discussion remains unresolved regarding the best practices for balancing these approaches.

Contextual Notes

Participants note the potential dangers of interpreting correlations without underlying theories and the complexities involved in statistical significance when multiple testing is considered.

Messages
15,681
Reaction score
10,475
  • Like
Likes   Reactions: RogueOne, Drakkith and Borg
Mathematics news on Phys.org
Indeed a very interesting article. Thanks for sharing it.
jedishrfu said:
We shouldn't look at data trying to find something interesting but should instead have a hypothesis in mind allowing the data to prove or disprove it and what can happen when we don't do that.
Yes, I completely agree with you.
 
There has been an ongoing discussion on the blog of Andrew Gelman, a professor of statistics at Columbia University, regarding p-hacking and deep data dives in general, and in particular the work of Brian Wansink, of which the Ars Technica article above refers to.

Here is one blog post, among many others:

http://andrewgelman.com/2016/12/15/hark-hark-p-value-heavens-gate-sings/
 
FiveThirtyEight also has run a few features about p-hacking. One has a nice interactive demonstrating how one can p-hack a dataset to say one conclusion or another (https://fivethirtyeight.com/features/science-isnt-broken/#part1) and in another, they run some surveys and p-hack to find spurious correlations such as those linking raw tomato consumption with Judaism or drinking lemonade with believing Crash deserved to win best picture (http://fivethirtyeight.com/features/you-cant-trust-what-you-read-about-nutrition/).

These are important points to consider when someone starts making wild claims about how data mining with artificial intelligence will revolutionize a new field or do something like help to cure cancer.
 
  • Like
Likes   Reactions: StatGuy2000 and RogueOne
In his book "Introduction to Medical Statistics" Second Edition Robert Mould gives a example of the dangers of interpreting correlations. Actual data on the number of storks documented in various towns shows a striking linear correlation with population. Finding correlations between seemingly unrelated variables can be dangerous in drawing conclusion if we do not have some underlying ideas for guidance of a possible relationship between the variables to start. In the case of the stork data a biologist would know that storks make nests on houses so no surprise with the stork population correlation.

P-hacking is statistics bass-ackwards.​
 
  • Like
Likes   Reactions: RogueOne
How do we reconcile the advice "Don't do p-hacking" with advice like "Always graph you data to see what it looks like"? Is this just a matter of accepting perceptions of patterns that we find visually "obvious" and rejecting patterns detected by other means?
 
Stephen Tashi said:
How do we reconcile the advice "Don't do p-hacking" with advice like "Always graph you data to see what it looks like"? Is this just a matter of accepting perceptions of patterns that we find visually "obvious" and rejecting patterns detected by other means?

I would say do the opposite of p-hacking. Analyze your data in multiple ways, and only trust your conclusion if the statistical significance is robust to multiple means of analysis.
 
Folks in data mining do a form of p hacking when scoring and clumping groups of data and they must develop a rationale that describes what they found.

As an example, analysis of bank customer history can identify a group of customers planning to leave the bank because they match others who have. From there you can drill down to see why both groups are similar and develop marketting plans to stem the loss.

In contrast, Cornell researchers developed a program to tease out the equations that describe a system based on measurement. It successfully discovered the equations of motion of a compound pendulum.

Some biology researchers did the same thing and got some great equatons but couldn't publish because they couldn't explain them with some new plausible theory.
 
jedishrfu said:
Folks in data mining do a form of p hacking when scoring and clumping groups of data and they must develop a rationale that describes what they found.
Therefore they adjust their significance levels for multiple testing.
 
  • #10
jedishrfu said:
An interesting article in Ars Technica on p-hacking vs deep data dives:

https://arstechnica.com/science/201...mindless-eating-mindless-research-is-bad-too/

We shouldn't look at data trying to find something interesting but should instead have a hypothesis in mind allowing the data to prove or disprove it and what can happen when we don't do that.
Of course you should look at data to find something interesting. The point is that you shouldn't use the same data to test your hypotheses.
 
  • Like
Likes   Reactions: Vanadium 50 and jedishrfu

Similar threads

  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 33 ·
2
Replies
33
Views
9K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 7 ·
Replies
7
Views
6K
  • · Replies 15 ·
Replies
15
Views
3K
  • · Replies 11 ·
Replies
11
Views
5K